Originally Posted by
birdspeed
The official report says the autopilot must have been off for the turn. As far as I know the only other report was from the IG who say they can replicate the turn by going to bank angle 25 to make the turn. I don’t understand or have seen a report using ADS-B as that was definitely not working during the turn….But also the continuing flight towards Penang is not straight, so the autopilot is looking like it is off here too. The most accurate study of the primary radar also shows a flight at no particular altitude(phugoid?).
So I would say it looks like the autopilot is not engaged.
With respect to the turn back, the Safety Investigation Report (SIR) states:
The reconstruction flight conducted on the B777 flight simulator had established that the turn back was likely made while the aircraft was under manual control and not the autopilot.
The IG found discrepancies between the radar position data and the ADS-B position data immediately before the transponder stopped transmitting. Those discrepancies called into question the accuracy of the turn entry point that was used in the simulator study. It was suggested that an earlier turn could be completed by the autopilot using a bank angle of 25°.
Regarding the military radar data, the SIR states:
It became very apparent, however, that the recorded altitude and speed change “blip” to “blip” were well beyond the capability of the aircraft. It was highlighted to the Team that the altitude and speed extracted from the data are subjected to inherent error. The only useful information obtained from the Military radar was the latitude and longitude position of the aircraft as this data is reasonably accurate.
The SIR also states:
The Team also noted that the aircraft’s flight path from after the turn was consistent with the navigation being set to LNAV and/or heading mode…
Which suggests the autopilot could well have been engaged during that part of the flight.